Interesting idea, but case not proved

BMJ 1997; 314 doi: (Published 01 February 1997) Cite this as: BMJ 1997;314:348
  1. Mike Pringle, professor of general practicea
  1. a Department of General Practice, Medical School, Queen's Medical Centre, Nottingham NG7 2UH


    B W Lloyd has produced a paper of interest and originality. If his evidence were compelling–or indeed if it were confirmed by others in a more rigorous study–then a small but important behavioural shift in doctor behaviour might follow. A more open consultation in which the patient feels respected as a full partner may benefit both parties. All my intuition (and my personal experience, for I too dictate many letters in front of patients) tells me that his conclusions are right. But intuition is clearly insufficient evidence especially in such an emotive area.

    Before I can be persuaded of Lloyd's conclusions I must firstly know much more about his study. There is implicit evidence that the setting was a paediatric clinic, but were the respondents to the questionnaires the patients, their parents, both, or a mixture? Were these consultations first attendances or follow ups, and if both, was the ratio the same in both groups?

    Perhaps the most important missing data relate to the case mix. Dictating a letter about asthma in front of the child and parents might be thought to be a low risk activity, but there are serious potential implications when the child has a new diagnosis of, say, leukaemia. If the lessons from this article are to be generalised, we need to know whether the general practice equivalent is dictating an antenatal referral letter in front of a newly pregnant woman, or dictating a letter to a psychiatrist. Indeed, no one is likely to advocate blanket policy of dictation in front of all patients; a selective approach would always be sensible.

    The second area that needs clarification, and further study, is the sensitivity of the research tool used. Only one pair of questions (“Did you like the doctor dictating the letter to the GP in front of you?” versus “Would you have liked to hear the doctor dictating his letter about your child to your GP?”) produced a difference between the two groups. However, one group is commenting on an experience received, while the other is speculating on an experience the nature of which it can only guess. It would have been astounding if such a loaded question did not yield a “positive” response.

    The other questions show a lack of awareness of the science of patient satisfaction. That all the respondents used only two of the four response categories shows that the survey was never likely to be a sensitive tool in this context. The use of tools validated in similar circumstances1 2 3 4 would be more likely to yield comparable and valid findings.

    It has to be acknowledged, however, that in a brief and provocative study Lloyd has posed a question that should stimulate others to explore this issue in greater depth; and that is presumably the challenge that the BMJ, by publishing this paper, is laying down to those of us interested in researching the consultation.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    View Abstract